Kb58's granite CNC router build

Bought two tool touch probes that will will help make up for not having an automatic tool changer (saving $1000, not including tool holders) by somewhat automating manual tool changes. The first probe is a movable unit that's used to set work piece position. The second probe is fixed in position off to the side of the work space. Once configured, the work piece is first located in X, Y, and Z with the movable probe, and then the job is started. When a tool change is needed, the router will stop the spindle, instruct what tool to install, then moves to the tool height sensor at the fixed location, measures and saves the tool height, then picks up where it left off. Doing the above by hand is slower and much more prone to mistakes or damage to the tool or work.
 
After being impressed with the VCarve Desktop demo, I bought it, and it wasn't much more expensive than MeshCAM. It has a huge customer base, which is good because there are a lot of YT videos explaining just about everything, and it's not subscription based. Everything is more or less intuitive and seems stable. Right now I'm learning how to use it using parts of the wooden gear clock plans I bought years ago but gave up on. This, after finding that my laser printer wasn't printing the paper templates at 100% the same in X and Y. After finding that I quit, because the gears would likely bind due to being elliptical. We'll see how this goes, starting with small simple parts.

Building wooden gear clocks will likely be a short adventure - I mean, how many inaccurate and short-running clocks does one need, when we all have access to atomic-based time via our phones? It's serving as the reward for teaching myself VCarve, and after that, who knows.
 
Spent the morning learning VCarve and ended up with a good portion of the clock parts - those most annoying to make by hand, like the gears - ready to go into the router. Haven't decided whether to start with good wood for the first go, or practice on something cheaper like low-quality plywood. Given how many false starts I had with the dust boot, the latter will probably win out.

I'm already wondering where in the house the clock will go, and how many times it'll be wound, but hey, it's kinetic art, not an accurate time-keeping device.
 
It be doing things! I was tickled watching the cutter almost magically turn raw material into precisely-machined shapes.

Started with a piece of MDF as a test piece. Part way through fabrication, I thought maybe it could serve as the frame, since it's plenty strong and the dark color contrasts nicely with the Baltic birch. It's 1/2" thick instead of 12 mm, 0.7 mm too thick (0.028"), and I don't know at this point whether that's an issue. I'll continue on, switching to 12 mm ply if necessary. Another way to correct it might be using the waste board surfacing cutter to remove the excess, but we'll see how this plays out.

Oh, and I can now say for a fact that a dust shoe is Not a nice-to-have - it's a must-have, at least when cutting MDF; the sawdust is like powdered cinnamon. The shoe isn't perfect; I could smell the MDF but couldn't see any on the work piece, though there was some lodged down in the cut slots. Overall, I'm extremely pleased with it, and it would have been a real mess in the garage without it.

Tomorrow I'll try cutting the majority of the gears in 9 mm plywood.

1721877525933.png
 
Cutting the 9 mm plywood went okay. It was about half done when I noticed that the cutter was starting to leave fuzzy/lumpy sidewalls on the cuts, but only on the second pass, so I'm guessing I was using the wrong feed or speed. The up/down-cutting 1/8" tool looked discolored good out toward the tip - overheated? Spindle speed was 16,000 rpm, which doesn't sound out of line. As for cleaning up the messy cut, VCarve Desktop doesn't have a "finish pass" feature, and while it can be faked, I wondered if rerunning the job with a new cutter would work. Other than the extra time, it worked well, but it's a reminder that I can't ignore feeds and speeds - yes I know better.

Oh, and I learned a tip about using painters tape and super glue. I didn't see a need to cover the entire panel with tape, so used only four stripes. Yeah well, that was all fine, until it was cutting parts between those stripes. It was nearing the end of cutting a small gear when I realized I better turn off the vacuum in case it broke free... too late, and with a "Thump" it disappeared down the vacuum hose, hah. That was fun because the vacuum has a bag, which had to be torn open to find the one valuable piece of "waste." Good times.

Somewhat related, I work on a project until I get tired of it, then switch to something different. Such was the case with the router, switching over to the belt sander to take a break. This is mentioned because there are several things that haven't been aligned on the router, ("how far off could it be?"). Well after cutting out all the clock parts, I decided on a whim to finally surface the waste board. When done, it was shocking to see a sawtooth pattern on the waste board that I could easily catch my fingernail on, which meant that the spindle wasn't oriented vertically. The mill tramming indicator confirmed that the spindle is leaning forward by about 1.5 mm over a distance of about 20 mm. For the clock, I'll "probably" get away with the misalignment because the cut spans are short enough that the cuts "probably" haven't exceeded tolerances. I think the easiest way is to shim the spindle mount itself. In addition to being off front-to-back, it's also off side-to-side, just to make things more fun.

1721877602873.png
 
This completes the build log, the router being completed in January 2024 - well, as "finished" as any homebuilt router ever is. Since then, three wooden gear clocks have been built and both the router and VCarve software have worked well. Ironically, after having built the router to someday machine aluminum, it may never happen. Not because it can't, but because I don't want to deal with chips flying across the garage. Oh yes, it's all doable - the dust shoe works better than I could have hoped for - but I'm not sure I want to. I have a mill that works for most machining, though if a project takes shape necessitating large curves, or multiple pieces, its ability to cut aluminum may be revisited.
 
durableoreo said:
I'm late to the party but I can't resist commenting. Welding ground precision rails onto a floppy piece of rectangular steel tube is... outrageous. Although it probably worked fine (I'm still reading the thread) and I bear you no ill will, even so, this an abomination. It's an "Illegal Operation", in the words of Robin Renzetti. Everything else so far looks great and thanks for posting the thread.
Click to expand...
No rails are welded anywhere on the project. I'm guessing what you saw are the rails sitting on cold-rolled steel bar that is, in turn, welded onto the gantry.
 
Last edited:
kb58 said:
No rails are welded anywhere on the project. I'm guessing what you saw are the rails sitting on cold-rolled steel bar that is, in turn, welded onto the gantry.
Click to expand...

Yes, I see it now. And there are plenty of other clues that I missed...

Such an impressive project. You've done all the things and machined granite besides. I'm assembling a small 12x12x12" fixed-gantry type mill at the moment and I hope I can see it through the way you've done with your build.
 
Thanks; it's just a hobby and was a challenge to see how good I could make one. I haven't used it in several months, due in part to my dear old dog passing away and getting a puppy that seems to consume as much time as a newborn child! In my spare time, CNC files are being prepared for clock number four. After that, who knows.
 
You must log in or register to reply here.
Back
Top